Detecting Floods for Disaster Relief
The model that will be created in this notebook can detect whether an area shown in an image is
flooded or not. The idea for creating this model has been spurred from the recent floodings in
Pakistan.
Such models can prove useful in flood relief, helping to detect which areas need immediate focus.
The dataset used to train this model is Louisiana flood 2016, uploaded by Kaggle user Rahul T P,
which you can view here.
The fastai library, a high level PyTorch library, has been used.
One of the points of this notebook is to showcase how simple it is to create powerful models. That
said, this notebook is not a tutorial or guide.
This model was trained on only 270 images and minimal code. Accessbility and abstraction to the
field of machine learning has come a long, long way. Given the right data and the right pretrained
model, a powerful model can be produced in less than an hour, if not half.
This is important: in disasters such as floods, the time taken to produce the logistics required for
relief can be drastically reduced. It is also important because the barrier of entry to this field is
dramatically lowered; more people can create powerful models, in turn producing better solutions.
https://www.kaggle.com/code/forbo7/detecting-floods-for-disaster-relief
Prediction on Disaster tweets using LSTM
https://www.kaggle.com/code/ryanchan911/prediction-on-disaster-tweets-using-lstm
Real-Time Climate Risk Mapping with BigQuery AI
AI-powered flood risk mapping system using BigQuery AI. The system processes unstructured
data — such as weather bulletins, social media signals, and real-time news reports — to
generate location-specific flood warnings with confidence scores.
Key Capabilities:
NLP + Embeddings: Transform unstructured text into risk indicators.
AI Risk Fusion: Combine weather anomalies, news sentiment, and historical floods.
Real-Time Dashboards: Deliver location-specific flood alerts and recommendations.
Cross-Border Coverage: United States & Canada with potential global scalability.
https://www.kaggle.com/code/aleenaazeem/real-time-climate-risk-mapping-with-bigquery-ai
Photographic Sources
https://earthobservatory.nasa.gov/images/50018/flood-extent-in-pakistan
Rainfall_1901_2016_PAK
The dataset contains 116 yeas of rainfall data in millimeters. The granularity is per month average
rainfall.
https://opendata.com.pk/dataset/rainfall-in-pakistan/resource/62edc0ad-f1e2-426d-944e-e3442c9e12a9